Thanks very much for your channel. It's one of my favorites. An aside, I hear you often doing what many others do, (including a close friend of mine who is an EE,) and refer to frequencies in 'hurt', whereas cycles per second are named after H. 'Hertz', so all frequencies should be spoken of as 'Hertz', not 'hurt'. I'm not sure what quirk of the mind causes so many to mispronounce the word so often, but there it is.
Hi John. Yes it's a bad habit of mine, I seem to know this but I it pops out when I am really trying to say Hertz, LOL. John Hurt, famous actor known for Alien, Heinrich Hertz, famous German physicist who discovered radio waves. I guess it's because "z" sounds like an "s" in English confusing a plural meaning. It's great to hear that you are enjoying the channel. I hope to have the next video uploaded in the next few days "Part 3 How to Build and Code a VFO". I have had some PCB's made up for the project which I plan to make available to people who wish to build the project, I am just waiting for the delivery. Best Regards, John M0XFX
I literally wasted *hours* yesterday when I realised something was *occasionally* a bit off off this this board. I was using the Adafruit library on an ESP32 but I think I'll switch to the EtherKit one now; thanks for the video! 👍 Liked & Subbed
Hi, Yes without calibration these boards certainly do have frequency stability issues. I can't comment on the Adafruit library as I have not tried it. The ESP32 is a great dev board and the Etherkit library will work fine on that platform. I make these videos through my own experiences in the hope that it's helps others. I am glad you found the video helpful and good luck with your project.
John, as soon as you integrate your real code, i.e not a calibration script it will change again. We should be integrating the calibration script into our main script. We forget that each line is one clock cycle. Not much of an issue on the lower bands but up to 10m and above it becomes an issue for sure. Depends how pedantic you are. Also, depending on the output drive you selected there will be more drift on the higher outputs for example 8ma for a simple mixer. They are not super stable at higher frequency, more down to design rather than function.
Hi, thank you for your comment. Yes that's correct, the correction needs to be implemented in your main script. In order to implement the cal figure into your project, add the following line in the code to the Arduino setup function. si5351.set_correction(-6190, SI5351_PLL_INPUT_XO); //Note: Change -6190 to the cal figure that you get from the cal script. The frequency shouldn't shift when changing the drive strength. If you are seeing change it may due to the harmonics and overtones coming out of the clock generator getting mixed up in your frequency counter. Adding a filter to clean up the clock signal may help. Also note: the ref crystals used in these breakout boards are not the best and can cause more noticeable instabilities at higher clock frequencies. I hope this is of help, regards John M0XFX
Hi, I think, that there is a better way to calibrate this device in 3 steps. 1. Set frequency to the nominal oscillator frequency like 25000000 2. Check the output frequency on frequency counter 3. Use this number as a second argument to the .si5351.init() function. Done
Excellent video, tutorial and effort to share = thank you! I am not a coder, at all, but I would think that a correction algorithm could be simply determined for each individual Si5351 unit and that it could be rather easily coded in... If the offset fault is linear over the entire bandwidth, then... Measure the offsets at the low and high ends of the range, subtract the low frequency from the high frequency, to determine the effected bandwidth... Divide the offset fault value by the effected bandwidth, to determine the correction factor, per Hertz. Code in this correction factor... Or, some other such means... I hope I said that clearly; it is very late at night. Anyway, cheers!
Thanks, John! I have a bad 20.2752 MHz oscillator in my TRS-80 Model 4 which nobody seems to make any more. I was looking at using this chip and learned about that easy to use library from this video. I was hoping a working sketch would fit in an attiny85... and... Miraculously, just setting the frequency with a correction factor comes to 8160 bytes... I'll have a whole 32 bytes of flash left. This means no bootloader, but that just means it'll start up faster. haha!
Hi, It's great to hear you have your TRS-80 back on frequency. Yes that's right if you only need a fixed frequency then just correction is that is needed. Thanks for the comment, best regards John M0XFX
I just bought an Si5351A clone on Amazon and it's not putting out a stable frequency. I'm quite new to this stuff so I have no clue what the problem is. The frequency is set to 10Mhz but it's being read by the frequency counter as 5.3/5.4Mhz. It jumps around quite a bit. I checked to see if I hooked something up wrong but everything seems to be connected properly. I'm guessing I have a bad module but I have limited equipment and limited knowledge so I'm hoping someone one here will have had the same issue and can point me to a solution.
Hi, The issue isn't the si5351 it's the crystal for the oscillator. You only have to carryout the calibration once to get the figure. Keep an eye on the channel as I am thinking of supply some SI5351 boards with the calibration figure so you don't need a frequency counter.
Haha, I'm sorry, you are correct. This has been a long-running joke with some of my viewers, playing on words involving John Hurt, the famous actor, and Heinrich Hertz, the physicist known for his work with electromagnetic waves. It all started with Michael Faraday and the unit of capacitance, the Farad, which is commonly referred to in plural as Farads. For example, people often say "10 microfarads." Should it be Farad or Farads? Hertz or Hertzs? I realise this can be confusing, so I will stop doing it. Thank you for pointing it out.
Hi, Yes the cal routine sketch should run on the ESP32 and will work as per the video. If you have an ESP32 you will need to configure the Arduino IDE to work with it. Keep an eye on the channel, I have a video coming soon on how to do this.
Thanks very much for your channel. It's one of my favorites.
An aside, I hear you often doing what many others do, (including a close friend of mine who is an EE,) and refer to frequencies in 'hurt', whereas cycles per second are named after H. 'Hertz', so all frequencies should be spoken of as 'Hertz', not 'hurt'. I'm not sure what quirk of the mind causes so many to mispronounce the word so often, but there it is.
Hi John. Yes it's a bad habit of mine, I seem to know this but I it pops out when I am really trying to say Hertz, LOL. John Hurt, famous actor known for Alien, Heinrich Hertz, famous German physicist who discovered radio waves. I guess it's because "z" sounds like an "s" in English confusing a plural meaning. It's great to hear that you are enjoying the channel. I hope to have the next video uploaded in the next few days "Part 3 How to Build and Code a VFO". I have had some PCB's made up for the project which I plan to make available to people who wish to build the project, I am just waiting for the delivery. Best Regards, John M0XFX
Awesome work. Can't wait to see part 3 of the series!
Thanks for comment, it's great to hear that you are enjoying the series. Part 3 should be out this coming weekend. Best Regards John
Hi John, Very good vid and information.
I literally wasted *hours* yesterday when I realised something was *occasionally* a bit off off this this board. I was using the Adafruit library on an ESP32 but I think I'll switch to the EtherKit one now; thanks for the video! 👍
Liked & Subbed
Hi, Yes without calibration these boards certainly do have frequency stability issues. I can't comment on the Adafruit library as I have not tried it. The ESP32 is a great dev board and the Etherkit library will work fine on that platform. I make these videos through my own experiences in the hope that it's helps others. I am glad you found the video helpful and good luck with your project.
John, as soon as you integrate your real code, i.e not a calibration script it will change again. We should be integrating the calibration script into our main script. We forget that each line is one clock cycle. Not much of an issue on the lower bands but up to 10m and above it becomes an issue for sure. Depends how pedantic you are.
Also, depending on the output drive you selected there will be more drift on the higher outputs for example 8ma for a simple mixer. They are not super stable at higher frequency, more down to design rather than function.
Hi, thank you for your comment. Yes that's correct, the correction needs to be implemented in your main script. In order to implement the cal figure into your project, add the following line in the code to the Arduino setup function.
si5351.set_correction(-6190, SI5351_PLL_INPUT_XO); //Note: Change -6190 to the cal figure that you get from the cal script.
The frequency shouldn't shift when changing the drive strength. If you are seeing change it may due to the harmonics and overtones coming out of the clock generator getting mixed up in your frequency counter. Adding a filter to clean up the clock signal may help. Also note: the ref crystals used in these breakout boards are not the best and can cause more noticeable instabilities at higher clock frequencies.
I hope this is of help, regards John M0XFX
Hi, I think, that there is a better way to calibrate this device in 3 steps.
1. Set frequency to the nominal oscillator frequency like 25000000
2. Check the output frequency on frequency counter
3. Use this number as a second argument to the .si5351.init() function.
Done
excellent John nice video well back to work
Excellent video, tutorial and effort to share = thank you! I am not a coder, at all, but I would think that a correction algorithm could be simply determined for each individual Si5351 unit and that it could be rather easily coded in... If the offset fault is linear over the entire bandwidth, then... Measure the offsets at the low and high ends of the range, subtract the low frequency from the high frequency, to determine the effected bandwidth... Divide the offset fault value by the effected bandwidth, to determine the correction factor, per Hertz. Code in this correction factor... Or, some other such means... I hope I said that clearly; it is very late at night. Anyway, cheers!
Thanks, John! I have a bad 20.2752 MHz oscillator in my TRS-80 Model 4 which nobody seems to make any more. I was looking at using this chip and learned about that easy to use library from this video. I was hoping a working sketch would fit in an attiny85... and... Miraculously, just setting the frequency with a correction factor comes to 8160 bytes... I'll have a whole 32 bytes of flash left. This means no bootloader, but that just means it'll start up faster. haha!
Hi, It's great to hear you have your TRS-80 back on frequency. Yes that's right if you only need a fixed frequency then just correction is that is needed. Thanks for the comment, best regards John M0XFX
Thank you !
Im on board!
Lovely work mate
John...how is that new DDS board you designed and showed a month or two going ?..any further developments?
Hi, Yes it's about there, I'm just waiting for the final PCB's to arrive. I will be making a video in the next few weeks.
@@M0XFXUK By the way I am enjoying this series you are doing now too.
I just bought an Si5351A clone on Amazon and it's not putting out a stable frequency. I'm quite new to this stuff so I have no clue what the problem is. The frequency is set to 10Mhz but it's being read by the frequency counter as 5.3/5.4Mhz. It jumps around quite a bit. I checked to see if I hooked something up wrong but everything seems to be connected properly. I'm guessing I have a bad module but I have limited equipment and limited knowledge so I'm hoping someone one here will have had the same issue and can point me to a solution.
✅✅✅😎
Very good nice calibration work 🎉
73z.📡
are all si5351 chips different to callibrate or can I just use your calibrating offset? because I dont have a frequency counter.
Hi, The issue isn't the si5351 it's the crystal for the oscillator. You only have to carryout the calibration once to get the figure. Keep an eye on the channel as I am thinking of supply some SI5351 boards with the calibration figure so you don't need a frequency counter.
@@M0XFXUKI, for one, would welcome that! I have 1 SI5351 Breakout board, and will be looking for another in the near future!
Nice demo, thank you. Sorry for being pedantic, but Hertz is the name of a person (Heinrich Hertz), it's not the plural of Hert. :)
Haha, I'm sorry, you are correct. This has been a long-running joke with some of my viewers, playing on words involving John Hurt, the famous actor, and Heinrich Hertz, the physicist known for his work with electromagnetic waves. It all started with Michael Faraday and the unit of capacitance, the Farad, which is commonly referred to in plural as Farads. For example, people often say "10 microfarads." Should it be Farad or Farads? Hertz or Hertzs? I realise this can be confusing, so I will stop doing it. Thank you for pointing it out.
Hi here... Can I use this example for an ESP32 DEV setup, if so what should I do? I am a bit of a novice in the field 🙂.
Hi, Yes the cal routine sketch should run on the ESP32 and will work as per the video. If you have an ESP32 you will need to configure the Arduino IDE to work with it. Keep an eye on the channel, I have a video coming soon on how to do this.
Thanks for the answer. so I'm looking forward to when you post it 🙂🙂.@@M0XFXUK